Accelerate Data Pipelines with Governance and Observability

Data teams are under more pressure than ever to ship faster, without cutting corners on quality or governance 📦. The problem? Too much time is still going to infrastructure work, brittle pipelines, and firefighting, instead of building the trusted data products the business actually needs. On April 2, Coalesce is joining Snowflake for a live 60-minute session on how #AI is changing the day-to-day reality of data engineering work. You'll see: 🔹 How AI can accelerate transformation logic, data modeling techniques, and pipeline development without removing human judgment from the equation 🔹 How to increase delivery velocity while keeping governance, lineage, and testing intact 🔹 Live demos from both Snowflake and Coalesce showing what this looks like in practice Join us for a practical look at how modern data pipeline development is evolving, with real examples you can take back to your team: https://lnkd.in/gbk7MVpS

Coalesce is teaming up with Snowflake for a session on how AI is reshaping day-to-day data engineering, helping teams move faster without compromising on quality or governance. Hope to see some of you there 👇

Most people still think of Snowflake as just a data warehouse. That’s outdated. Snowflake has quietly evolved into a platform where Data + AI + LLMs live together natively. With features like: • LLMs directly callable via SQL (Cortex) • Native vector embeddings for semantic search • Built-in support for RAG architectures • No data movement, fully governed AI You can now build: 1. Chatbots on enterprise data 2. Document intelligence pipelines 3. Semantic search systems All inside Snowflake. The biggest gap right now is not capability… it’s awareness. I wrote a blog breaking down Snowflake’s GenAI & LLM capabilities in a simple, practical way - no hype, just real understanding. https://lnkd.in/gaT7Cmwc If you’re working with data, this shift is something you don’t want to miss. #Snowflake #GenAI #LLM #DataEngineering #AI #MachineLearning #DataPlatforms

Thanks Sanket Prabhu for this great writeup on Snowflake's GenAI & LLM capabilities — worth a read for anyone building on the modern data stack. Beyond the well-known data governance and security story, what genuinely impresses me about Snowflake as a data platform: 𝘛𝘩𝘪𝘯𝘬𝘪𝘯𝘨 𝘣𝘢𝘤𝘬 𝘵𝘰 𝘸𝘩𝘦𝘯 𝘐 𝘧𝘪𝘳𝘴𝘵 𝘴𝘵𝘢𝘳𝘵𝘦𝘥 𝘶𝘴𝘪𝘯𝘨 𝘚𝘯𝘰𝘸𝘧𝘭𝘢𝘬𝘦 𝘪𝘯 2024 — 𝘵𝘩𝘦 𝘔𝘓 𝘔𝘰𝘥𝘦𝘭 𝘙𝘦𝘨𝘪𝘴𝘵𝘳𝘺 𝘩𝘢𝘥 𝘫𝘶𝘴𝘵 𝘭𝘢𝘶𝘯𝘤𝘩𝘦𝘥. 𝘞𝘩𝘢𝘵'𝘴 𝘴𝘵𝘶𝘤𝘬 𝘸𝘪𝘵𝘩 𝘮𝘦 𝘴𝘪𝘯𝘤𝘦 𝘪𝘴 𝘩𝘰𝘸 𝘤𝘰𝘯𝘴𝘪𝘴𝘵𝘦𝘯𝘵𝘭𝘺 𝘚𝘯𝘰𝘸𝘧𝘭𝘢𝘬𝘦 𝘱𝘢𝘤𝘬𝘢𝘨𝘦𝘴 𝘢𝘥𝘷𝘢𝘯𝘤𝘦𝘥 𝘧𝘶𝘯𝘤𝘵𝘪𝘰𝘯𝘢𝘭𝘪𝘵𝘺 𝘣𝘦𝘩𝘪𝘯𝘥 𝘢 𝘚𝘘𝘓 𝘪𝘯𝘵𝘦𝘳𝘧𝘢𝘤𝘦. 𝘝𝘦𝘤𝘵𝘰𝘳 𝘴𝘦𝘢𝘳𝘤𝘩, 𝘔𝘓 𝘪𝘯𝘧𝘦𝘳𝘦𝘯𝘤𝘦, 𝘓𝘓𝘔 𝘤𝘢𝘭𝘭𝘴 — 𝘢𝘭𝘭 𝘢𝘤𝘤𝘦𝘴𝘴𝘪𝘣𝘭𝘦 𝘷𝘪𝘢 𝘚𝘘𝘓. 𝘛𝘩𝘢𝘵'𝘴 𝘢 𝘮𝘦𝘢𝘯𝘪𝘯𝘨𝘧𝘶𝘭 𝘶𝘯𝘭𝘰𝘤𝘬: 𝘣𝘶𝘴𝘪𝘯𝘦𝘴𝘴 𝘶𝘴𝘦𝘳𝘴 𝘢𝘯𝘥 𝘢𝘯𝘢𝘭𝘺𝘴𝘵𝘴 𝘤𝘢𝘯 𝘵𝘢𝘱 𝘪𝘯𝘵𝘰 𝘎𝘦𝘯𝘈𝘐 𝘤𝘢𝘱𝘢𝘣𝘪𝘭𝘪𝘵𝘪𝘦𝘴 𝘸𝘪𝘵𝘩𝘰𝘶𝘵 𝘴𝘸𝘪𝘵𝘤𝘩𝘪𝘯𝘨 𝘵𝘰𝘰𝘭𝘴 𝘰𝘳 𝘭𝘦𝘢𝘳𝘯𝘪𝘯𝘨 𝘯𝘦𝘸 𝘱𝘢𝘳𝘢𝘥𝘪𝘨𝘮𝘴. 𝘌𝘷𝘦𝘯 𝘰𝘱𝘦𝘳𝘢𝘵𝘪𝘰𝘯𝘢𝘭 𝘵𝘢𝘴𝘬𝘴 𝘭𝘪𝘬𝘦 𝘴𝘤𝘢𝘭𝘪𝘯𝘨 𝘢𝘯𝘥 𝘳𝘦𝘴𝘰𝘶𝘳𝘤𝘦 𝘮𝘢𝘯𝘢𝘨𝘦𝘮𝘦𝘯𝘵 𝘢𝘳𝘦 𝘢 𝘲𝘶𝘦𝘳𝘺 𝘢𝘸𝘢𝘺. And through all the feature releases, Snowsight remains genuinely clean and intuitive. That's harder to maintain than it looks. The SQL-first philosophy isn't just a convenience — it's an architectural choice that keeps data governance, lineage, and access control consistent across traditional pipelines and AI workloads alike. Highly recommend Sanket's article if you want a grounded view of where Snowflake's GenAI story actually stands today. 👇 And please — don't be misled by posts still calling Snowflake "just a cloud data warehouse." That take is really, really old. The platform has moved far beyond that, and articles like Sanket's are a good reminder of how far.

Coding agents offered by data platforms are not moats. They are an interface layer. Their value primarily depends on whether they help you unlock the differentiated capabilities of the underlying platform. That is why the real question is not whether a platform has a coding agent but whether that agent helps users get more value out of what the platform already does well. In Databricks' case, Genie Code is valuable because it helps customers quickly build pipelines using Lakeflow, dashboards with AI/BI Dashboards, operational databases with Lakebase, and more. So, with that said, here is a level-headed discussion without fluffy vendor hype with Weston Hutchins, Director of Product @ Databricks, answering key questions around Genie Code. Enjoy!

Perficient has earned the Cortex Code Preferred Partner Badge from Snowflake — a recognition based on demonstrated impact. Our teams use Cortex Code, Snowflake’s AI-powered coding agent, to streamline engineering, machine learning, and application development workflows. It supports data engineering, machine learning, and application delivery across active Snowflake projects, with more than 3,000 real requests so far. This badge reflects proven enterprise adoption, platform expertise, successful implementations, and a commitment to sharing what we’ve learned with the broader community. It highlights teams who are already applying AI‑assisted development in meaningful, repeatable ways. Clients deserve partners with hands‑on experience, not just an interest in emerging tools. This recognition reinforces the work our teams are already delivering on Snowflake. Read how we earned this badge and what it means for your Snowflake initiatives! https://lnkd.in/e3NDrMvq #SnowflakePartner #CortexCode #EnterpriseAI #GenAI #DataAI

I was the beta tester assigned to Coco’s ML skills earlier this year. I can say with total confidence that they are top-tier skills for training, evaluating and deploying ML models. The results were shocking. I gave Coco a slate of 56 separate ML tasks from basic model fitting, to model eval tracking to full blown container service model and feature registry deployments with inferenecing endpoints and eval tracking applications. Coco aced every test. I honestly don’t know what else to say. I’ve been an MLE, I’ve worked as a Data Scientist it a top shop, I’ve seen and done it all. Coco can do it too. Just try it. 🤷

Cortex Code (CoCo) is proving to be much more than an AI sidekick—it’s the foundation for Agentic ML on Snowflake.

Just used Snowflake's Cortex Code (CoCo) to build and deploy an application in a container all inside Snowflake. No separate dev environment. No external hosting. The AI agent understands your data context, generates the code, and deploys it through Snowpark Container Services. Still early. Not everything was seamless. But the direction is clear the line between "where your data lives" and "where you build" is disappearing. #Snowflake #CortexCode #AI #DataEngineering #Analytics